xdi message

Subject: More very interesting $has-related questions

From: Drummond Reed <drummond.reed@xdi.org>
To: OASIS - XDI TC <xdi@lists.oasis-open.org>
Date: Thu, 1 Apr 2010 01:37:40 -0700

This is related to the whole question about when/if it matters what $has relationships are expressed as XDI predicates vs. subjects.

Again, this only comes up because we're now saying $has statements can be expressed either way, i.e., =example/+email and =example+email identify the same point in the XDI graph.

So the first question is: does an XDI $get for the XRI =example/$has and and XDI $get for the XRI =example$has return the same results? In other words, are they equivalent statements?

The second question is: if they are equivalent, does that mean that an XDI processor should consider all three of the XDI graphs below equivalent?

#1 FULLY COLLAPSED

�

=example

�� $is$a

�� +person

�� +email

�� (/+home+email)

�� (/+work+email)

�� +home+email

�� (/+email$1)

�� +work+email

�� (/+email$1)

�� +email$1

�� example@example.com�

�

#2 ALL BUT LAST SUBSEGMENT COLLAPSED

�

=example

�� $is$a

�� +person

�� $has

�� +home

�� +work

�� +email

=example+home

�� +email

�� (=example+email/$1)

=example+work

�� +email

�� (=example+email/$1)

=example+email

�� $1

�� example@example.com�

�

#3 FULLY EXPANDED

�

=example

�� $is$a

�� +person

�� $has

�� +home+email

�� +work+email

�� +email$1

=example+home+email

�� $is

�� =example+email$1

=example+work+email

�� $is

�� =example+email$1

=example+email$1

�� $value

�� example@example.com�

*********************
Again, this will make for fascinating discussion on tomorrow's call.

=Drummond